File size: 23,862 Bytes
9ff1a64 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 |
2023-10-17 10:57:18,983 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,984 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 10:57:18,984 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,984 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 Train: 966 sentences
2023-10-17 10:57:18,985 (train_with_dev=False, train_with_test=False)
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 Training Params:
2023-10-17 10:57:18,985 - learning_rate: "3e-05"
2023-10-17 10:57:18,985 - mini_batch_size: "8"
2023-10-17 10:57:18,985 - max_epochs: "10"
2023-10-17 10:57:18,985 - shuffle: "True"
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 Plugins:
2023-10-17 10:57:18,985 - TensorboardLogger
2023-10-17 10:57:18,985 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:57:18,985 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 Computation:
2023-10-17 10:57:18,985 - compute on device: cuda:0
2023-10-17 10:57:18,985 - embedding storage: none
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 Model training base path: "hmbench-ajmc/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:18,985 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:57:19,714 epoch 1 - iter 12/121 - loss 4.11176953 - time (sec): 0.73 - samples/sec: 3310.46 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:57:20,438 epoch 1 - iter 24/121 - loss 3.76005368 - time (sec): 1.45 - samples/sec: 3272.87 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:57:21,175 epoch 1 - iter 36/121 - loss 3.10341988 - time (sec): 2.19 - samples/sec: 3232.18 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:57:21,939 epoch 1 - iter 48/121 - loss 2.54022751 - time (sec): 2.95 - samples/sec: 3306.53 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:57:22,623 epoch 1 - iter 60/121 - loss 2.20458174 - time (sec): 3.64 - samples/sec: 3312.27 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:57:23,405 epoch 1 - iter 72/121 - loss 1.91912829 - time (sec): 4.42 - samples/sec: 3309.16 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:57:24,148 epoch 1 - iter 84/121 - loss 1.68059317 - time (sec): 5.16 - samples/sec: 3353.48 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:57:24,937 epoch 1 - iter 96/121 - loss 1.49481711 - time (sec): 5.95 - samples/sec: 3380.76 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:57:25,639 epoch 1 - iter 108/121 - loss 1.38776546 - time (sec): 6.65 - samples/sec: 3359.76 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:57:26,354 epoch 1 - iter 120/121 - loss 1.28622241 - time (sec): 7.37 - samples/sec: 3340.58 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:57:26,401 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:26,401 EPOCH 1 done: loss 1.2827 - lr: 0.000030
2023-10-17 10:57:27,280 DEV : loss 0.2441699206829071 - f1-score (micro avg) 0.5721
2023-10-17 10:57:27,288 saving best model
2023-10-17 10:57:27,722 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:28,445 epoch 2 - iter 12/121 - loss 0.20649691 - time (sec): 0.72 - samples/sec: 3182.29 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:57:29,175 epoch 2 - iter 24/121 - loss 0.22995980 - time (sec): 1.45 - samples/sec: 3218.68 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:57:29,928 epoch 2 - iter 36/121 - loss 0.23059976 - time (sec): 2.20 - samples/sec: 3223.00 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:57:30,663 epoch 2 - iter 48/121 - loss 0.22267191 - time (sec): 2.94 - samples/sec: 3238.25 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:57:31,424 epoch 2 - iter 60/121 - loss 0.21515131 - time (sec): 3.70 - samples/sec: 3306.79 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:57:32,133 epoch 2 - iter 72/121 - loss 0.21250351 - time (sec): 4.41 - samples/sec: 3298.44 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:57:32,892 epoch 2 - iter 84/121 - loss 0.20931862 - time (sec): 5.17 - samples/sec: 3316.16 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:57:33,676 epoch 2 - iter 96/121 - loss 0.20235956 - time (sec): 5.95 - samples/sec: 3306.08 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:57:34,437 epoch 2 - iter 108/121 - loss 0.20444909 - time (sec): 6.71 - samples/sec: 3296.50 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:57:35,196 epoch 2 - iter 120/121 - loss 0.19934455 - time (sec): 7.47 - samples/sec: 3287.52 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:57:35,266 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:35,266 EPOCH 2 done: loss 0.1981 - lr: 0.000027
2023-10-17 10:57:36,024 DEV : loss 0.13820908963680267 - f1-score (micro avg) 0.8161
2023-10-17 10:57:36,030 saving best model
2023-10-17 10:57:36,558 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:37,315 epoch 3 - iter 12/121 - loss 0.10671627 - time (sec): 0.75 - samples/sec: 3199.00 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:57:38,087 epoch 3 - iter 24/121 - loss 0.10895843 - time (sec): 1.53 - samples/sec: 3252.71 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:57:38,888 epoch 3 - iter 36/121 - loss 0.09820473 - time (sec): 2.33 - samples/sec: 3307.37 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:57:39,619 epoch 3 - iter 48/121 - loss 0.09907103 - time (sec): 3.06 - samples/sec: 3313.97 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:57:40,388 epoch 3 - iter 60/121 - loss 0.10159016 - time (sec): 3.83 - samples/sec: 3279.99 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:57:41,125 epoch 3 - iter 72/121 - loss 0.10049162 - time (sec): 4.56 - samples/sec: 3290.90 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:57:41,905 epoch 3 - iter 84/121 - loss 0.10314644 - time (sec): 5.34 - samples/sec: 3293.60 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:57:42,586 epoch 3 - iter 96/121 - loss 0.10316924 - time (sec): 6.03 - samples/sec: 3247.80 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:57:43,425 epoch 3 - iter 108/121 - loss 0.10749021 - time (sec): 6.86 - samples/sec: 3250.57 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:57:44,135 epoch 3 - iter 120/121 - loss 0.10975500 - time (sec): 7.57 - samples/sec: 3241.35 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:57:44,190 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:44,190 EPOCH 3 done: loss 0.1105 - lr: 0.000023
2023-10-17 10:57:44,981 DEV : loss 0.1287086457014084 - f1-score (micro avg) 0.7931
2023-10-17 10:57:44,987 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:45,701 epoch 4 - iter 12/121 - loss 0.09747709 - time (sec): 0.71 - samples/sec: 2907.36 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:57:46,425 epoch 4 - iter 24/121 - loss 0.09056427 - time (sec): 1.44 - samples/sec: 3172.85 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:57:47,181 epoch 4 - iter 36/121 - loss 0.07817966 - time (sec): 2.19 - samples/sec: 3174.55 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:57:47,980 epoch 4 - iter 48/121 - loss 0.07983687 - time (sec): 2.99 - samples/sec: 3194.07 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:57:48,746 epoch 4 - iter 60/121 - loss 0.07961880 - time (sec): 3.76 - samples/sec: 3252.18 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:57:49,457 epoch 4 - iter 72/121 - loss 0.08005860 - time (sec): 4.47 - samples/sec: 3224.86 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:57:50,244 epoch 4 - iter 84/121 - loss 0.08381911 - time (sec): 5.26 - samples/sec: 3199.52 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:57:51,052 epoch 4 - iter 96/121 - loss 0.07951076 - time (sec): 6.06 - samples/sec: 3191.68 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:57:51,880 epoch 4 - iter 108/121 - loss 0.08050723 - time (sec): 6.89 - samples/sec: 3178.37 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:57:52,607 epoch 4 - iter 120/121 - loss 0.07858401 - time (sec): 7.62 - samples/sec: 3232.92 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:57:52,656 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:52,656 EPOCH 4 done: loss 0.0783 - lr: 0.000020
2023-10-17 10:57:53,416 DEV : loss 0.1368216872215271 - f1-score (micro avg) 0.8428
2023-10-17 10:57:53,421 saving best model
2023-10-17 10:57:53,955 ----------------------------------------------------------------------------------------------------
2023-10-17 10:57:54,692 epoch 5 - iter 12/121 - loss 0.05030366 - time (sec): 0.73 - samples/sec: 2978.83 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:57:55,441 epoch 5 - iter 24/121 - loss 0.05842618 - time (sec): 1.48 - samples/sec: 3004.98 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:57:56,202 epoch 5 - iter 36/121 - loss 0.05402638 - time (sec): 2.24 - samples/sec: 3143.73 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:57:56,977 epoch 5 - iter 48/121 - loss 0.05579888 - time (sec): 3.01 - samples/sec: 3107.83 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:57:57,688 epoch 5 - iter 60/121 - loss 0.05433023 - time (sec): 3.73 - samples/sec: 3159.77 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:57:58,487 epoch 5 - iter 72/121 - loss 0.05588435 - time (sec): 4.52 - samples/sec: 3222.11 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:57:59,183 epoch 5 - iter 84/121 - loss 0.06059224 - time (sec): 5.22 - samples/sec: 3228.21 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:57:59,937 epoch 5 - iter 96/121 - loss 0.05994432 - time (sec): 5.98 - samples/sec: 3237.32 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:58:00,703 epoch 5 - iter 108/121 - loss 0.05885631 - time (sec): 6.74 - samples/sec: 3244.65 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:58:01,507 epoch 5 - iter 120/121 - loss 0.05832928 - time (sec): 7.54 - samples/sec: 3267.73 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:58:01,553 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:01,553 EPOCH 5 done: loss 0.0585 - lr: 0.000017
2023-10-17 10:58:02,302 DEV : loss 0.16064168512821198 - f1-score (micro avg) 0.8162
2023-10-17 10:58:02,307 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:03,052 epoch 6 - iter 12/121 - loss 0.04426736 - time (sec): 0.74 - samples/sec: 3294.18 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:58:03,802 epoch 6 - iter 24/121 - loss 0.05380387 - time (sec): 1.49 - samples/sec: 3310.51 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:58:04,550 epoch 6 - iter 36/121 - loss 0.05088120 - time (sec): 2.24 - samples/sec: 3350.87 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:58:05,236 epoch 6 - iter 48/121 - loss 0.04892439 - time (sec): 2.93 - samples/sec: 3321.57 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:58:06,003 epoch 6 - iter 60/121 - loss 0.04484164 - time (sec): 3.70 - samples/sec: 3304.78 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:58:06,692 epoch 6 - iter 72/121 - loss 0.04516285 - time (sec): 4.38 - samples/sec: 3254.88 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:58:07,506 epoch 6 - iter 84/121 - loss 0.04649830 - time (sec): 5.20 - samples/sec: 3282.01 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:58:08,260 epoch 6 - iter 96/121 - loss 0.04408763 - time (sec): 5.95 - samples/sec: 3282.93 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:58:09,055 epoch 6 - iter 108/121 - loss 0.04327841 - time (sec): 6.75 - samples/sec: 3250.65 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:58:09,841 epoch 6 - iter 120/121 - loss 0.04469431 - time (sec): 7.53 - samples/sec: 3261.23 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:58:09,899 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:09,899 EPOCH 6 done: loss 0.0444 - lr: 0.000013
2023-10-17 10:58:10,661 DEV : loss 0.16509920358657837 - f1-score (micro avg) 0.8256
2023-10-17 10:58:10,666 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:11,437 epoch 7 - iter 12/121 - loss 0.01541501 - time (sec): 0.77 - samples/sec: 3108.65 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:58:12,150 epoch 7 - iter 24/121 - loss 0.02410835 - time (sec): 1.48 - samples/sec: 3142.06 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:58:12,860 epoch 7 - iter 36/121 - loss 0.02628934 - time (sec): 2.19 - samples/sec: 3297.76 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:58:13,611 epoch 7 - iter 48/121 - loss 0.02740439 - time (sec): 2.94 - samples/sec: 3283.05 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:58:14,402 epoch 7 - iter 60/121 - loss 0.03048279 - time (sec): 3.73 - samples/sec: 3282.57 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:58:15,133 epoch 7 - iter 72/121 - loss 0.03201688 - time (sec): 4.47 - samples/sec: 3263.94 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:58:15,878 epoch 7 - iter 84/121 - loss 0.03254549 - time (sec): 5.21 - samples/sec: 3247.21 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:58:16,610 epoch 7 - iter 96/121 - loss 0.03124162 - time (sec): 5.94 - samples/sec: 3271.43 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:58:17,415 epoch 7 - iter 108/121 - loss 0.03286000 - time (sec): 6.75 - samples/sec: 3266.85 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:58:18,164 epoch 7 - iter 120/121 - loss 0.03208405 - time (sec): 7.50 - samples/sec: 3286.59 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:58:18,219 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:18,219 EPOCH 7 done: loss 0.0320 - lr: 0.000010
2023-10-17 10:58:18,971 DEV : loss 0.18511991202831268 - f1-score (micro avg) 0.8542
2023-10-17 10:58:18,976 saving best model
2023-10-17 10:58:19,628 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:20,370 epoch 8 - iter 12/121 - loss 0.03326804 - time (sec): 0.74 - samples/sec: 2896.45 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:58:21,124 epoch 8 - iter 24/121 - loss 0.02660413 - time (sec): 1.49 - samples/sec: 3211.26 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:58:21,883 epoch 8 - iter 36/121 - loss 0.02728278 - time (sec): 2.25 - samples/sec: 3177.93 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:58:22,704 epoch 8 - iter 48/121 - loss 0.02521362 - time (sec): 3.07 - samples/sec: 3259.47 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:58:23,482 epoch 8 - iter 60/121 - loss 0.02360167 - time (sec): 3.85 - samples/sec: 3252.99 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:58:24,230 epoch 8 - iter 72/121 - loss 0.02285615 - time (sec): 4.60 - samples/sec: 3257.76 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:58:24,950 epoch 8 - iter 84/121 - loss 0.02475812 - time (sec): 5.32 - samples/sec: 3227.04 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:58:25,699 epoch 8 - iter 96/121 - loss 0.02481499 - time (sec): 6.07 - samples/sec: 3234.77 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:58:26,456 epoch 8 - iter 108/121 - loss 0.02381609 - time (sec): 6.83 - samples/sec: 3260.39 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:58:27,226 epoch 8 - iter 120/121 - loss 0.02539417 - time (sec): 7.60 - samples/sec: 3239.88 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:58:27,278 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:27,278 EPOCH 8 done: loss 0.0255 - lr: 0.000007
2023-10-17 10:58:28,030 DEV : loss 0.1753813475370407 - f1-score (micro avg) 0.8515
2023-10-17 10:58:28,035 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:28,780 epoch 9 - iter 12/121 - loss 0.02319572 - time (sec): 0.74 - samples/sec: 3358.61 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:58:29,544 epoch 9 - iter 24/121 - loss 0.02035831 - time (sec): 1.51 - samples/sec: 3182.04 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:58:30,299 epoch 9 - iter 36/121 - loss 0.01757965 - time (sec): 2.26 - samples/sec: 3098.64 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:58:31,049 epoch 9 - iter 48/121 - loss 0.01978687 - time (sec): 3.01 - samples/sec: 3092.72 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:58:31,770 epoch 9 - iter 60/121 - loss 0.01965078 - time (sec): 3.73 - samples/sec: 3098.63 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:58:32,538 epoch 9 - iter 72/121 - loss 0.01917681 - time (sec): 4.50 - samples/sec: 3162.47 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:58:33,313 epoch 9 - iter 84/121 - loss 0.02104908 - time (sec): 5.28 - samples/sec: 3188.02 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:58:34,078 epoch 9 - iter 96/121 - loss 0.01936059 - time (sec): 6.04 - samples/sec: 3228.86 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:58:34,903 epoch 9 - iter 108/121 - loss 0.01966485 - time (sec): 6.87 - samples/sec: 3233.29 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:58:35,661 epoch 9 - iter 120/121 - loss 0.01996074 - time (sec): 7.62 - samples/sec: 3226.11 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:58:35,713 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:35,713 EPOCH 9 done: loss 0.0202 - lr: 0.000004
2023-10-17 10:58:36,476 DEV : loss 0.18414919078350067 - f1-score (micro avg) 0.8446
2023-10-17 10:58:36,481 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:37,192 epoch 10 - iter 12/121 - loss 0.01422985 - time (sec): 0.71 - samples/sec: 3331.69 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:58:37,934 epoch 10 - iter 24/121 - loss 0.02050312 - time (sec): 1.45 - samples/sec: 3144.41 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:58:38,683 epoch 10 - iter 36/121 - loss 0.01665361 - time (sec): 2.20 - samples/sec: 3252.90 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:58:39,417 epoch 10 - iter 48/121 - loss 0.01607792 - time (sec): 2.93 - samples/sec: 3302.06 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:58:40,247 epoch 10 - iter 60/121 - loss 0.01888085 - time (sec): 3.76 - samples/sec: 3376.90 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:58:40,984 epoch 10 - iter 72/121 - loss 0.01759198 - time (sec): 4.50 - samples/sec: 3302.85 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:58:41,737 epoch 10 - iter 84/121 - loss 0.01703542 - time (sec): 5.26 - samples/sec: 3248.04 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:58:42,542 epoch 10 - iter 96/121 - loss 0.01766957 - time (sec): 6.06 - samples/sec: 3251.25 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:58:43,353 epoch 10 - iter 108/121 - loss 0.01802283 - time (sec): 6.87 - samples/sec: 3264.13 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:58:44,058 epoch 10 - iter 120/121 - loss 0.01776160 - time (sec): 7.58 - samples/sec: 3243.89 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:58:44,116 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:44,116 EPOCH 10 done: loss 0.0176 - lr: 0.000000
2023-10-17 10:58:44,870 DEV : loss 0.19578655064105988 - f1-score (micro avg) 0.8375
2023-10-17 10:58:45,260 ----------------------------------------------------------------------------------------------------
2023-10-17 10:58:45,261 Loading model from best epoch ...
2023-10-17 10:58:46,626 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-17 10:58:47,299
Results:
- F-score (micro) 0.8071
- F-score (macro) 0.5559
- Accuracy 0.6974
By class:
precision recall f1-score support
pers 0.8194 0.8489 0.8339 139
scope 0.8626 0.8760 0.8692 129
work 0.6596 0.7750 0.7126 80
loc 1.0000 0.2222 0.3636 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7951 0.8194 0.8071 360
macro avg 0.6683 0.5444 0.5559 360
weighted avg 0.7971 0.8194 0.8009 360
2023-10-17 10:58:47,299 ----------------------------------------------------------------------------------------------------
|